The scoring cutover is long done (infrastructure/clickhouse/MIGRATION.md: "Status (2026-07-09) — the cutover is complete and the legacy engine is gone") and the GENESIS quest went public around 2026-07-16. Two artefacts of that period are still in the tree:

genesis_public — a feature flag that is now permanently on, still threaded through 8 gate sites, an admin page, and 6 test files.
The legacy transcription tooling — the one-shot machinery that moved pre-cutover scores out of Mongo. It is unwired dead code.

Together they also keep alive a documented naming hazard: seasons/genesis/implementation.md section 0 exists solely to warn that the scoring cutover genesis baseline and the GENESIS quest campaign are unrelated and "must not share identifiers".

WARNING: the flag cannot simply be deleted

genesis_public.enabledAt is the SOLE source of genesisEpoch() (lib/app/feature-flags.ts). Deleting the flag document deletes the epoch, and genesisEpoch() returning null does not mean "no gate" — it means COUNT LIFETIME ACTIVITY. The consequences:

Every operative is instantly regrandfathered to a high tier, directly contradicting the from-zero design of 3689d1f6 (feat(quest): GENESIS cutover — from-zero at the flip, no grandfathering, #897).
swamp.quest_signal_counts bakes the epoch into the materialized view WHERE clause. See infrastructure/clickhouse/CLAUDE.md: "The exact epoch is genesis_public.enabledAt (UTC ms), not the flip day." App-side lifetime counts and ClickHouse-side post-epoch counts would then disagree permanently.

So the GATE is vestigial, but the STAMP is load-bearing. Retiring the flag is a migration, not a delete.

Proposed work

Phase 1 — delete the dead transcription scaffolding (safe now, no migration)

lib/repositories/legacy-genesis.ts — the LegacyGenesisRow / LegacyGenesisArchive / GenesisTranscriptionGateway ports.
lib/infrastructure/clickhouse/clickhouse-genesis-gateway.ts — ClickHouseGenesisGateway. VERIFIED UNWIRED: a repo-wide grep across lib/, routes/, scripts/ and tests/ finds only the class definition itself. It is not in the composition root.
infrastructure/clickhouse/genesis-transcription.sql — labelled "Run ONCE at cutover". Already run.
infrastructure/clickhouse/init/04-genesis.sql — creates the STAGING table swamp.legacy_genesis_input, only ever populated by the now-deleted exporter.
The historical runbook in infrastructure/clickhouse/MIGRATION.md, which already labels itself "not a live dependency".
Orphaned comments in lib/infrastructure/mongo-metric-snapshot-repository.ts describing how it fed the deleted exporter.

Worth stating as motivation rather than a footnote: the gateway docstring claims a lockstep test enforces byte-equivalence between its embedded SQL and the .sql file. Both lib/app/legacy-genesis-export.ts and tests/infrastructure/genesis_transcription_lockstep_test.ts NO LONGER EXIST (verified). The file therefore documents a guarantee that nothing provides — it is actively misleading, not merely unused.

Phase 2 — retire genesis_public (needs the epoch pinned first)

Read the production value of feature_flags.genesis_public.enabledAt and pin it as a constant, or via the existing GENESIS_STARTS_AT escape hatch already parsed by parseGenesisEpoch in lib/quest-catalog.ts. It must be the EXACT instant, not the flip day.
Point genesisEpoch() at that constant and delete the flag read path, the 30s TTL cache, isGenesisVisible, setGenesisPublic and setGenesisPublicForTests.
Remove the 8 gate sites: routes/leaderboard.tsx, routes/api/quest/genesis/claim.ts, routes/api/quest/genesis/ghost.ts, routes/api/u/[username]/genesis.ts, routes/u/[username]/genesis.tsx, routes/u/[username].tsx, routes/api/v1/u/[username]/card.ts, lib/app/quest/announce-quest-tier.ts.
Remove routes/admin/genesis.tsx and its tab in components/admin/AdminTabs.tsx.
Update the 6 test files that pin the flag.

Interaction with #1055 (just merged): quest_completion stores epochMs and treats a changed epoch as a seal invalidation. If Phase 2 pins a different instant than the flag currently yields, every sealed pass invalidates once. That is harmless — one recompute wave, which is simply the pre-1055 behaviour — but it should be a known consequence rather than a surprise, and it is a good reason to pin the exact stored value.

Explicit non-goals — do NOT delete these

Everything below is either live product or live data, and shares the word genesis only by the collision that section 0 warns about:

The GENESIS quest / battle pass — roughly 5,000 lines, live since around 2026-07-16: /quest, /u/NAME/genesis, the Season Journey card, the claim ceremony, swamp quest, the public card.
grant_type=genesis rows and their read path — the frozen pre-cutover score history for every operative. lib/infrastructure/clickhouse/clickhouse-score-reads.ts sums them in the live board/score/graph read. Deleting them zeroes every historical score.
SCORE_GRANTS_CUTOVER — live config that pins the activity projector so it never double-counts what the genesis grants already hold. Still validated at telemetry boot.
lib/app/genesis-leaderboard.ts and components/GenesisBoard.tsx — despite the name this is the QUEST tier board (it reads Mongo badge_awards), rendered on /leaderboard and /quest. tests/app/genesis_leaderboard_test.ts covers this, not the scoring baseline.

Suggested sequencing

Phase 1 is independent and can land on its own: it is pure deletion of unreachable code and carries no runtime risk. Phase 2 should be a separate PR precisely because it is a behaviour-preserving migration that only LOOKS like a deletion, and it deserves its own review.

02Bog Flow

Open

8/11/2026, 12:06:24 AM

No activity in this phase yet.

03Sludge Pulse

data.latest() is memoized per run, contradicting the documented "sees current-run data" contract

Lab Insights not publicly available

Model lookup misreports EMFILE (too many open files) as a broken definition file and dumps a raw stacktrace

Client shows 'Authentication failed' when actually rate-limited

Workers retry forever on permanent auth failures instead of backing off

Docs: add swamp access group list-idp to access commands reference

serve telemetry identity warnings are diagnostic but not actionable — say how to fix them

swamp quest: drop the heartbeat field the server no longer sends

--grant-reload auto does not detect filesystem changes — blind to ConfigMap updates

Server token definitions are not durable across pod restarts

WebSocket auth rate limiter blocks legitimate users when sharing IP with a failing client

Add native Bedrock Knowledge Base Retrieve support

GC expired OAuth server tokens and control plane secrets

migrate-index silently drops entries with shallow paths — all of workflows-evaluated/ and any root-level file

Cluster health snapshots with SSE streaming for swamp serve

Add swamp access group list-idp to enumerate IdP groups seen by the server

Grants directory loader skips symlinks — breaks Kubernetes ConfigMap mounts

Daemon telemetry flush: stop() can skip its final flush when a timer drain is already in flight

Support --grants-dir for loading grants from a directory

Docs: `swamp compile` subcommand doesn't exist

Guard the projector's full rewrites against the fail-open lease (ADV-6 from #1575/#1576)

Projector write amplification: board_totals/owner_devices full rewrites are owners × runs/day

Profile candidateIds prune costs a full ledger scan to build itself — 42.88M rows for one keyed read

Leaderboard phase 3: owner-grain board_totals so a cache miss is survivable at 10B

swamp serve is telemetry-dark: scheduled workflow runs never reach /ingest, and the daemon never flushes

CLI has never sent insert_id: retried telemetry batches are double-counted into cli_daily, inflating XP in an append-only ledger

Catalog backfill reconciles against the local cache, not the datastore

Extension bundle cache not invalidated when an imported sibling module changes (entry-file fingerprint only)

swamp serve: data.query fails permanently with "UNIQUE constraint failed: catalog.*" — one duplicate catalog 5-tuple aborts every catalog rebuild

dailyHistory range pushdown — deferred from #1576, and the measurement says why

score-rollup projector runs N× per cadence — the run gate is per-replica memory, not shared state

Docs: manual never mentions telemetry, and the serve/scheduling pages imply automated runs are visible when they are not

Explore: CLI binary update integrity lacks independent verification

Leaderboard phase 2: take score_grants off the board read path — read today from score_daily

Explore: reconcile quality rubric's dormant provenance factor

Explore: pull-time signature verification for extensions

Explore: transparency-log integration (Rekor) for extension signatures

Explore: publish-time artifact signing for extensions (cosign keyless)

Profile day drill-down returns 0 entries at the 5s ClickHouse ceiling for every operative

A ledger read failure still renders as 'no activity': loadProfileScore swallows to an empty standing behind HTTP 200

reconcile_from_disk_bench.ts fails deno check (stale DenoRuntime stub)

Improve multi-word search query handling and guidance

Retire the genesis_public flag and delete the dead cutover transcription scaffolding

Leaderboard phase 1: cache the board read and bust it on tier crossings (no schema change)

Catalog backfill deletes rows for any model whose data names are numeric

Docs: document catalog completeness check and --repair in doctor reference

Leaderboard board windows re-anchor on wall-clock while the rest of the view is frozen at asOf

Decision: move per-event grant MVs to day-grain — ledger is on track for ~1.5B score_grants / ~4B quest_grants

Leaderboard phase 5: score-rollup-projector write cost at 10B — ~430M rows scanned per 60s tick

Epic: /leaderboard and /u/{username} do not survive 10B events — measured audit + phased plan

Datastore-managed configuration for multi-instance serve deployments

Ghosts on the leaderboard 500

swamp command doesn't check cwd

swamp auth login process fails when run remotely

serve boot OOM: findAllGlobalSince loads all workflow runs as full aggregates during reconciliation

extension search: channel-filtered results carry null version fields, and multiple --channel values collapse instead of unioning

s3-datastore: localHasAllRemoteEntries uses bare rel instead of localRelPath, always returns false in namespaced mode

datastoreSetupExtension migrates .swamp/ to cache root instead of namespace directory

swamp serve writes outputs/ outside the bound namespace — silently never persisted to the remote datastore

s3-datastore: bulk pull writes to cache root instead of the namespace, wedging push and arming a full-datastore delete

git: add pull, fetch --tags, and cherry-pick methods

Workflow runs with wide dataArtifact fan-out (~400/run) OOM both workflow history get and swamp serve boot, crash-looping the scheduler; rollback doesn't help (state-triggered, not release-triggered)

Show user:<name> instead of user:<id> in grant list output

Reusable long-lived invite link: one URL to bring people into the swamp

Exclude swamp internal model types from model list output

Skill docs use invalid two-argument 'vault get' form

access token mint prints an invalid retrieval command (vault get, not vault read-secret)

serve reload and model edit timeout over --server in Kubernetes

swamp auth whoami should report paid tier and trial status

Use whoami entitlement to explain private-extension push refusals

Document --restricted-commands flag for swamp serve

vault put: pass tags through to extension provider CreateSecret/setSecret calls

Docs: update manual for 23 new --server commands

Update VaultProvider interface in @systeminit/swamp-testing to support optional tags in put()

Docs: document --grants-file flag for swamp serve

Docs: document --server support for serve reload

Docs: add --tag flag to vault put reference

Docs: update manual for --detach-runs deprecation and HA auto-detection

Document the Codex plugin in swamp docs

Add --server support to CLI commands missing it